Patient Preference Research With AI-Lowered Evidence Generation
Author(s)
Emanuele Pietropaolo, PhD1, Matteo Bracco, MSc1, Martina Di Blasio, MSc1, Rosanna Comoretto, PhD1, Giuseppe Rizzo, PhD2, Alessia Visconti, PhD1, Ileana Baldi, PhD3, PAOLA BERCHIALLA, PhD1.
1University of Torino, Torino, Italy, 2Links Foundation, Torino, Italy, 3University of Padova, Padova, Italy.
1University of Torino, Torino, Italy, 2Links Foundation, Torino, Italy, 3University of Padova, Padova, Italy.
OBJECTIVES: This work aims to examine the persistent gaps in the availability, accessibility, and use of Patient Preference Information (PPI) within the healthcare decision-making landscape. Despite increasing recognition of the value of patient perspectives in clinical research, regulatory evaluation, and therapeutic development, there remains no standardized approach for identifying Patient Preference Studies (PPS) or a centralized, up-to-date repository to support evidence generation. The objective is to explore these challenges and outline conceptual directions for improving the integration of PPI into health-related decision processes.
METHODS: Repertor.IO started as an academically funded project based on advanced AI technologies to automate and accelerate the processes of retrieval, organization, and analysis of PPS. It employs embeddings-based indexing to store articles according to their semantic content, enabling contextual and efficient searches. The platform automated data extraction capabilities allows for delivering accurate data extraction while minimizing time and errors. Furthermore, advanced meta-analytical capabilities identify methodologically compatible studies, supporting quantitative syntheses of PPI.
RESULTS: Repertor.IO offers an innovative online platform designed to streamline the use of PPSs through four core capabilities: i) a continuously updated structured database that identifies, aggregates, categorize and extracts key data from PPSs; ii) AI integration for intuitive navigation and contextual search of relevant information; iii) automated data extraction and download functions for instant access to essential study details; and iv) meta-analytic evidence generation through ad-hoc statistical models and AI driven integration. Together these features significantly reduce the research time and enhance the applicability of Patient Preference evidence across the medical product life cycle.
CONCLUSIONS: Repertor.IO enables researchers to quickly access and review PPSs, using the most recent evidence, and identify gaps in the existing disease areas. Combined with robust statistical analysis of quantitative PPI, these capabilities support and facilitate formal evidence-based decision-making and promote more structured, informed processes.
METHODS: Repertor.IO started as an academically funded project based on advanced AI technologies to automate and accelerate the processes of retrieval, organization, and analysis of PPS. It employs embeddings-based indexing to store articles according to their semantic content, enabling contextual and efficient searches. The platform automated data extraction capabilities allows for delivering accurate data extraction while minimizing time and errors. Furthermore, advanced meta-analytical capabilities identify methodologically compatible studies, supporting quantitative syntheses of PPI.
RESULTS: Repertor.IO offers an innovative online platform designed to streamline the use of PPSs through four core capabilities: i) a continuously updated structured database that identifies, aggregates, categorize and extracts key data from PPSs; ii) AI integration for intuitive navigation and contextual search of relevant information; iii) automated data extraction and download functions for instant access to essential study details; and iv) meta-analytic evidence generation through ad-hoc statistical models and AI driven integration. Together these features significantly reduce the research time and enhance the applicability of Patient Preference evidence across the medical product life cycle.
CONCLUSIONS: Repertor.IO enables researchers to quickly access and review PPSs, using the most recent evidence, and identify gaps in the existing disease areas. Combined with robust statistical analysis of quantitative PPI, these capabilities support and facilitate formal evidence-based decision-making and promote more structured, informed processes.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
PCR177
Topic
Methodological & Statistical Research, Patient-Centered Research, Real World Data & Information Systems
Topic Subcategory
Health State Utilities
Disease
No Additional Disease & Conditions/Specialized Treatment Areas